Vous êtes sur la page 1sur 8

J. Nat. Prod.

2000, 63, 1035-1042 1035

Flavonoids as Antioxidants

Pier-Giorgio Pietta*
Institute of Advanced Biomedical Technologies, National Council of Research Via F.lli Cervi 93, 20090 Segrate (MI), Italy

Received September 13, 1999

Flavonoids are phenolic substances isolated from a wide range of vascular plants, with over 8000 individual
compounds known. They act in plants as antioxidants, antimicrobials, photoreceptors, visual attractors,
feeding repellants, and for light screening. Many studies have suggested that flavonoids exhibit biological
activities, including antiallergenic, antiviral, antiinflammatory, and vasodilating actions. However, most
interest has been devoted to the antioxidant activity of flavonoids, which is due to their ability to reduce
free radical formation and to scavenge free radicals. The capacity of flavonoids to act as antioxidants in
vitro has been the subject of several studies in the past years, and important structure-activity
relationships of the antioxidant activity have been established. The antioxidant efficacy of flavonoids in
vivo is less documented, presumably because of the limited knowledge on their uptake in humans. Most
ingested flavonoids are extensively degraded to various phenolic acids, some of which still possess a radical-
scavenging ability. Both the absorbed flavonoids and their metabolites may display an in vivo antioxidant
activity, which is evidenced experimentally by the increase of the plasma antioxidant status, the sparing
effect on vitamin E of erythrocyte membranes and low-density lipoproteins, and the preservation of
erythrocyte membrane polyunsaturated fatty acids. This review presents the current knowledge on
structural aspects and in vitro antioxidant capacity of most common flavonoids as well as in vivo
antioxidant activity and effects on endogenous antioxidants.

Introduction it, such as heart disease, cataracts, cognitive dysfunction,

and cancer.4-6 Humans have evolved with antioxidant
Oxidation is the transfer of electrons from one atom to
systems to protect against free radicals. These systems
another and represents an essential part of aerobic life and
include some antioxidants produced in the body (endog-
our metabolism, since oxygen is the ultimate electron
acceptor in the electron flow system that produces energy enous) and others obtained from the diet (exogenous). The
in the form of ATP.1 However, problems may arise when first include (a) enzymatic defenses, such as Se-glutathione
the electron flow becomes uncoupled (transfer of unpaired peroxidase, catalase, and superoxide dismutase, which
single electrons), generating free radicals. Examples of metabolize superoxide, hydrogen peroxide, and lipid per-
oxygen-centered free radicals, known as reactive oxygen oxides, thus preventing most of the formation of the toxic
species (ROS), include superoxide (O2•-), peroxyl (ROO•), HO•, and (b) nonenzymatic defenses, such as glutathione,
alkoxyl (RO•), hydroxyl (HO•), and nitric oxide (NO•). The histidine-peptides, the iron-binding proteins transferrin
hydroxyl (half-life of 10-9 s) and the alkoxyl (half-life of and ferritin, dihydrolipoic acid, reduced CoQ10, melatonin,
seconds) free radicals are very reactive and rapidly attack urate, and plasma protein thiols, with the last two ac-
the molecules in nearby cells, and probably the damage counting for the major contribution to the radical-trapping
caused by them is unavoidable and is dealt with by repair capacity of plasma. The various defenses are complemen-
processes. On the other hand, the superoxide anion, lipid tary to each other, since they act against different species
hydroperoxides, and nitric oxide are less reactive.2 In at different cellular compartments. However, despite these
addition to these ROS radicals, in living organisms there defense antioxidants (able either to suppress free radical
are other ROS nonradicals, such as the singlet oxygen (1O2), formation and chain initiation or to scavenge free radical
hydrogen peroxide (H2O2), and hypochlorous acid (HOCl). and chain propagation), some ROS still escape to cause
It is accepted that ROS play different roles in vivo. Some damage. Thus, the body antioxidant system is provided also
are positive and are related to their involvement in energy by repair antioxidants (able to repair damage, and based
production, phagocytosis, regulation of cell growth and on proteases, lipases, transferases, and DNA repair en-
intercellular signaling, and synthesis of biologically im- zymes).7
portant compounds.3 However, ROS may be very damaging, Owing to the incomplete efficiency of our endogenous
since they can attack lipids in cell membranes, proteins in defense systems and the existence of some physiopatho-
tissues or enzymes, carbohydrates, and DNA, to induce logical situations (cigarette smoke, air pollutants, UV
oxidations, which cause membrane damage, protein modi- radiation, high polyunsaturated fatty acid diet, inflamma-
fication (including enzymes), and DNA damage. This tion, ischemia/reperfusion, etc.) in which ROS are produced
oxidative damage is considered to play a causative role in in excess and at the wrong time and place, dietary
aging and several degenerative diseases associated with antioxidants are needed for diminishing the cumulative
effects of oxidative damage over the life span.8,9 Well-
* To whom correspondence should be addressed. Tel.: 0039 02 26422725. established antioxidants derived from the diet are vitamins
Fax: 0039 02 26422770. E-mail: pietta@itba.mi.cnr.it. C, E, A, and carotenoids, which have been studied inten-
10.1021/np9904509 CCC: $19.00 © 2000 American Chemical Society and American Society of Pharmacognosy
Published on Web 05/27/2000
1036 Journal of Natural Products, 2000, Vol. 63, No. 7

ROS produced by the photosynthetic electron transport

system.20 Furthermore, because of their favorable UV-
absorbing properties, flavonoids protect plants from UV
radiation of sun and scavenge UV-generated ROS.21
Apart from their physiological roles in the plants, fla-
vonoids are important components in the human diet,
although they are generally considered as nonnutrients.
Figure 1. Basic flavonoid structure.
Indeed, the level of intake of flavonoids from diet is
considerably high as compared to those of vitamin C (70
mg/day), vitamin E (7-10 mg/day), and carotenoids (β-
sively.10 Besides these antioxidant vitamins, other sub-
carotene, 2-3 mg/day).22 Flavonoid intake can range
stances in plants might account for at least part of the
between 50 and 800 mg/day, depending on the consumption
health benefits associated with vegetable and fruit con-
of vegetables and fruit, and of specific beverages, such as
sumption. Over the past decade evidence has been ac-
red wine, tea, and unfiltered beer.23 In particular, red wine
cumulated that plant polyphenols are an important class
and tea contain high levels (approximately 200 mg per
of defense antioxidants. These compounds are widespread
glass of red wine or cup of tea) of total phenols. Thus,
virtually in all plant foods, often at high levels, and include
variations in consumption of these beverages are mainly
phenols, phenolic acids, flavonoids, tannins, and lignans.
responsible for the overall flavonoid intake in different
Flavonoids. Flavonoids are formed in plants from the
national diets. Another significant source of flavonoids are
aromatic amino acids phenylalanine and tyrosine, and
different medicinal plants and related phytomedicines.24
malonate.11 The basic flavonoid structure is the flavan
nucleus, which consists of 15 carbon atoms arranged in Epidemiological Evidence. Several epidemiological
three rings (C6-C3-C6), which are labeled A, B, and C studies provide support for a protective effect of the
(Figure 1). The various classes of flavonoids differ in the consumption of fresh fruits and vegetables against can-
level of oxidation and pattern of substitution of the C ring, cer,25,26 heart disease,27-29 and stroke.30,31 Normally, high
while individual compounds within a class differ in the consumers of fruits and vegetables have a healthy lifestyle,
pattern of substitution of the A and B rings. Among the which may be an important factor for their resistance
many classes of flavonoids, those of particular interest to against chronic diseases. All in all, fruits and vegetables
this review are flavones, flavanones, isoflavones, flavonols, do play a preventive role, which is due to a variety of
flavanonols, flavan-3-ols, and anthocyanidins (Table 1). constituents, including vitamins, minerals, fiber, and nu-
Other flavonoid classes include biflavones, chalcones, au- merous phytochemicals, including flavonoids. Thus, it is
rones, and coumarins. Hydrolyzable tannins, proantho- possible that also flavonoids contribute to the protective
cyanidins (flavan-3-ol oligomers), caffeates, and lignans are effect of fruits and vegetables. This possibility has been
all plant phenols, and they are usually classified separately. evidenced by several in vitro, ex vivo, and animal studies.32
Flavonoids generally occur in plants as glycosylated Unfortunately, the evidence in humans is still limited and
derivatives, and they contribute to the brilliant shades of somewhat controversial.33 Data on biological markers, such
blue, scarlet, and orange, in leaves, flowers, and fruits.12 as blood levels of flavonoids and their metabolites, are not
Apart from various vegetables and fruits, flavonoids are widely available, thus making it difficult to determine the
found in seeds, nuts, grains, spices, and different medicinal individual or the combined role of the flavonoids and other
plants as well in beverages, such as wine (particularly red antioxidants.
wine), tea, and (at lower levels) beer.13 More specifically, The association between flavonoid intake and cancer
the flavones apigenin and luteolin are common in cereal protection is (at present) weak. According to some epi-
grains and aromatic herbs (parsley, rosemary, thyme), demiological studies, there is no evidence that flavonoid
while their hydrogenated analogues hesperetin and nar- intake is protective against some types of cancer.34 Only
ingin are almost exclusively present in citrus fruits.14 The one study has shown that the consumption of flavonoids
flavonols quercetin and kaempferol are predominant in is inversely correlated with lung cancer.35 In contrast, a
vegetables and fruits, where they are found mainly in the possible protective role against coronary heart disease of
skin, with the exception of onions. Isoflavones are found flavonoid intake (either from fruits and vegetables or red
most often in legumes, including soybeans, black beans, wine and tea) has been reported in four out of six
green beans, and chick peas. Alfalfa and clover sprouts and epidemiological studies.36 The dietary sources of flavonoids
sunflower seeds also contain isoflavones.15 The flavan-3- were fruits, vegetables, red wine, and tea, and they were
ols (+)-catechin, (-)-epicatechin, (-)-epigallocatechin, and found to be inversely correlated with the risk of coronary
their gallate esters are widely distributed in plants, heart disease and stroke. On the other hand, a weak
although they are very rich in tea leaves. Flavan oligomers inverse relationship was observed by Knekt and co-work-
(proanthocyanidins) are present in apples, grapes, berries, ers,37 and in the largest prospective cohort study conducted
persimmon, black currant, and sorghum and barley grains.16 in the United States38 only a weak but nonsignificant
Anthocyanidins and their glycosides (anthocyanins) are inverse correlation was observed for flavonoid consumption
natural pigments and are abundant in berries and red and coronary mortality. Thus, it appears that the effects
grape.17 of flavonoids are strongest for coronary heart disease
Flavonoids play different roles in the ecology of plants. mortality and not morbidity. Accordingly, the present
Due to their attractive colors, flavones, flavonols, and epidemiological data (although far from conclusive) evi-
anthocyanidins may act as visual signals for pollinating dence a possible protective role of dietary flavonoids, thus
insects. Because of their astringency, catechins and other making desirable a regular consumption of foods and
flavanols can represent a defense system against insects beverages rich in flavonoids.
harmful to the plant.18 Flavonoids act as catalysts in the In Vitro Antioxidant Action. According to Halliwell
light phase of photosynthesis and/or as regulators of iron and Gutteridge,6 mechanisms of antioxidant action can
channels involved in phosphorylation.19 They can also include (1) suppressing reactive oxygen species formation
function as stress protectants in plant cells by scavenging either by inhibition of enzymes or chelating trace elements
Journal of Natural Products, 2000, Vol. 63, No. 7 1037
Table 1. Structures of Flavonoids Table 1 (Continued)

Flavones Flavan-3-ols
5 7 3′ 4′ 3 5 7 3′ 4′ 5′
luteolin OH OH OH OH (+)-catechin βOH OH OH OH OH
apigenin OH OH OH (-)-epicatechin ROH OH OH OH OH
chrysin OH OH (-)-epigallocatechin ROH OH OH OH OH OH

Flavylium Salts
3 5 7 3′ 4′
5 7 3′ 4′
cyanidin OH OH OH OH OH
hesperetin OH OH OH OCH3 cyanin O-glc OH OH OH OH
naringenin OH OH OH pelargonidin OH OH OH - OH

involved in free radical production; (2) scavenging reactive

oxygen species; and (3) upregulating or protecting antioxi-
dant defenses.
Flavonoids have been identified as fulfilling most of the
criteria described above. Thus, their effects are twofold.
1. Flavonoids inhibit the enzymes responsible for super-
oxide anion production, such as xanthine oxidase39 and
Flavonols protein kinase C.40 Flavonoids have been also shown to
5 7 3′ 4′ 5′ inhibit cyclooxygenase, lipoxygenase, microsomal monooxy-
quercetin OH OH OH OH
genase, glutathione S-transferase, mitochondrial succin-
kaempferol OH OH OH oxidase, and NADH oxidase, all involved in reactive oxygen
galangin OH OH species generation.41,42
fisetin OH OH OH A number of flavonoids efficiently chelate trace metals,
myricetin OH OH OH OH OH which play an important role in oxygen metabolism. Free
iron and copper are potential enhancers of reactive oxygen
species formation, as exemplified by the reduction of
hydrogen peroxide with generation of the highly aggressive
hydroxyl radical,

H2O2 + Fe2+(Cu+) f •OH + OH- + Fe3+(Cu2+)

or by the copper-mediated LDL (low-density lipoprotein)

Flavanonol oxidation,
5 7 3′ 4′
taxifolin OH OH OH OH LH f L• f LOO•

where LH represents LDL.

Nevertheless, it has to be remembered that these metal
ions are essential for many physiological functions, as
constituents of hemoproteins and cofactors of different
enzymes, including those involved (iron for catalase, copper
for ceruloplasmin and Cu,Zn-superoxide dismutase) in the
Isoflavones antioxidant defense.43
5 7 4′ The proposed binding sites for trace metals to flavonoids
are the catechol moiety in ring B, the 3-hydroxyl, 4-oxo
genistein OH OH OH
genistin OH Oglc OH
groups in the heterocyclic ring, and the 4-oxo, 5-hydroxyl
daidzein OH OH groups between the heterocyclic and the A rings (Figure
daidzin Oglc OH 2). However, the major contribution to metal chelation is
biochanin A OH OH OCH3 due to the catechol moiety, as exemplified by the more
formononetin OH OCH3 pronounced bathochromic shift produced by chelation of
1038 Journal of Natural Products, 2000, Vol. 63, No. 7

Table 2. One-Electron Reduction Potentials at pH 7 (E7, V) for

Selected Radical Couples (Adapted from Ref 46)
HO•, H+/HO 2.310
RO•, H+/ROH (alkoxyl) 1.600
ROO•, H+/ROOH (peroxyl) 1.000
PUFA•, H+/PUFA-H 0.600
HU•, H+/UH2 (urate) 0.590
TO•, H+/TOH 0.480
ascorbate•-, H+/ascorbate- 0.282

2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), an

aqueous soluble vitamin E analogue. The Trolox equivalent
Figure 2. Binding sites for trace metals. antioxidant capacity (TEAC) is defined as the concentration
of Trolox with the same antioxidant capacity as a 1 mM
concentration of the antioxidant under investigation. Jo-
vanovic et al.45 adopted the same approach described for
determining the priority46 (hierarchy of one-electron reduc-
tion potentials at pH 7, Table 2) of radical couples and
evaluated the in vitro antioxidant potential of flavonoids
on the basis of the one-electron reduction potential at pH
7 (E7) of the Fl-O•/Fl-OH pair. In contrast, the half-peak
oxidation potentials (Ep/2) of flavonoids have been proposed
as suitable parameters to evaluate the scavenging activity.
This assumes that both the electrochemical oxidation Fl-
OH f Fl-O• + e- + H+ and the hydrogen atom donating
reaction Fl-OH f Fl-O• + H• involve the breaking of the
Figure 3. Scavenging of ROS (R•) by flavonoids. same O-H bond.54 According to this approach, flavonoids
with Ep/2 < 0.2 are defined as readily oxidable and
copper to quercetin compared to that of kaempferol (similar therefore good scavengers. TEAC, E7, and Ep/2 values of
in structure to quercetin except that it lacks the catechol selected flavonoids are compared in Table 3.
group in the B ring).44 The data obtained by the three different approaches
2. Due to their lower redox potentials (0.23 < E7 < 0.75 mentioned above provide clear evidence that the radical-
V),45 flavonoids (Fl-OH) are thermodynamically able to scavenging activity depends on the structure and the
reduce highly oxidizing free radicals with redox potentials substituents of the heterocyclic and B rings, as suggested
in the range 2.13-1.0 V,46 such as superoxide, peroxyl, by Bors et al.55 More specifically, the major determinants
alkoxyl, and hydroxyl radicals by hydrogen atom donation: for radical-scavenging capability are (i) the presence of a
catechol group in ring B, which has the better electron-
Fl-OH + R• f Fl-O• + RH donating properties and is a radical target, and (ii) a 2,3-
double bond conjugated with the 4-oxo group, which is
where R• represents superoxide anion, peroxyl, alkoxyl, and responsible for electron delocalization.
hydroxyl radicals.47-49 The presence of a 3-hydroxyl group in the heterocyclic
The aroxyl radical (Fl-O•) may react with a second ring also increases the radical-scavenging activity, while
radical, acquiring a stable quinone structure (Figure 3). additional hydroxyl or methoxyl groups at positions 3,5 and
The aroxyl radicals could interact with oxygen, generat- 7 of rings A and C seem to be less important. These
ing quinones and superoxide anion, rather than terminat- structural features contribute to increase the stability of
ing chain reactions. The last reaction may take place in the aroxyl radical, i.e., the antioxidant capacity of the
the presence of high levels of transient metal ions and is parent flavonoid. Thus, flavonols and flavones containing
responsible for the undesired prooxidant effect of fla- a catechol group in ring B are highly active, with flavonols
vonoids.50 Thus, the overall capacity of flavonoids to act more potent than the corresponding flavones because of the
as antioxidants depends not only on the redox potential of presence of the 3-hydroxyl group. Glycosylation of this
the couple Fl-O•/Fl-OH but also on possible side reactions group, as in rutin, reduces greatly the radical-scavenging
of the aroxyl radical. Scavenging of superoxide is particu- capacity. An additional hydroxyl group in ring B (pyrogallol
larly important, because this radical is ubiquitous in group) enhances further the antioxidant capacity, as
aerobic cells and, despite its mild reactivity, is a potential exemplified by myricetin. On the contrary, the presence of
precursor of the aggressive hydroxyl radical in the Fenton only one hydroxyl in ring B diminishes the activity.
and Haber-Weiss reactions.51 Besides scavenging, fla- Flavanonols and flavanones, due to the lack of conjugation
vonoids may stabilize free radicals involved in oxidative provided by the 2,3-double bond with the 4-oxo group, are
processes by complexing with them.52 weak antioxidants. Antioxidant activities of flavan mono-
Many studies have been performed to establish the mers are comparable to those of flavanonols (catechin vs
relationship between flavonoid structure and their radical- taxifolin). However, the presence of a pyrogallol group in
scavenging activity, and the most relevant are briefly ring B (like in epigallocatechin) or galloylation of the
described. Rice-Evans et al.53 developed a valuable assay 3-hydroxyl group (as for epigallocatechin gallate and epi-
that allows for the determination of the hierarchy of catechin gallate) enhances the antioxidant capacity.
radical-scavenging ability of flavonoids (and related phe- Anthocyanidins and their glycosides (anthocyanins) are
nolic acids). This assay is based on the ability of an equipotent to quercetin and catechin gallates, provided that
antioxidant to scavenge (at pH 7.4) a preformed radical a catechol structure is present in ring B (like in cyanidin).
cation chromophore of 2,2′-azinobis(3-ethylbenzothiazoline- Removal of the 3-hydroxyl group from ring B, as in
6-sulfonic acid (ABTS•+) in relation to that of 6-hydroxy- pelargonidin, reduces the antioxidant capacity at the same
Journal of Natural Products, 2000, Vol. 63, No. 7 1039
Table 3. Trolox Equivalent Antioxidant Capacity (mM), E7 (V), Table 4. Trolox Equivalent Antioxidant Capacity (mM) of
and Ep/2 (V) of Flavonoids (the Hydroxylation Pattern Is Shown Isoflavones and Selected Flavonoid Metabolites (the
in Parentheses for Each Component) Hydroxylation Pattern Is Shown in Parentheses for Each
TEAC (mM) E7 (V) Ep/2 (V) genistein (5, 7, 4′) 2.90
quercetin (3, 5, 7, 3′, 4′) 4.7 0.33 0.06 biochanin A (5, 7, 4′OCH3) 1.16
quercetin 3-rutinoside 2.42 0.6 0.18 daidzein (7, 4′) 1.25
kaempferol (3, 5, 7, 4′) 1.34 0.75 0.12 formononetin (7, 4′OCH3) 0.11
myricetin (3, 5, 7, 3′, 4′, 5′) 3.10 0.36 genistein 7-glucoside 1.24
galangin (3, 5, 7) 1.49 0.62 0.32
Selected Flavonoid Metabolites
Flavanonols 3,4-dihydroxyphenylacetic acid (I) 2.16
3-methoxy-4-hydroxyphenylacetic acid (II) 1.63
TEAC (mM) E7 (V) Ep/2 (V) 3-methoxy-4-hydroxybenzoic acid (III) 1.19
taxifolin (3, 5, 7, 3′, 4′) 1.9 0.5 0.15 3,4-dihydroxybenzoic acid (IV) 1.01
dihydrokaempferol (3, 5, 7, 3′) 1.39 3-methoxy-4-hydroxyhippuric acid (V) 1.29

fication of LDL in vitro, provided that they are added before
TEAC (mM) E7(V) Ep/2(V) the initiation of the oxidation.61 What is interesting is that
luteolin (5, 7, 3′, 4′) 2.09 0.6 0.18 the most active flavonoids possess the same structural
luteolin 4′-glucoside 1.74 features that guarantee TEAC efficacy or low redox poten-
apigenin (5, 7, 4′) 1.45 >1 tials.
chrysin (5, 7) 1.43 >1 In Vivo Flavonoid Antioxidant Potential. Despite
Flavanones the increasing evidence for the in vitro antioxidant poten-
TEAC (mM) E7 (V) Ep/2 (V) tial of flavonoids, little is known about their efficacy in vivo,
and this may be ascribed to the only sketchy knowledge
eriodictyol (5, 7, 3′, 4′) 1.8 on their bioavailability in humans. Only recently has it
hesperetin [5, 7, 3′, 4′(och3)] 1.4 0.4
naringenin (5, 7, 4′) 1.5 0.6 been proved that flavonoids from dietary sources are
naringenin 7-rutinoside 0.8 absorbed at an extent that may promote an antioxidant
effect. According to most authors, flavonol, flavone, and
Catechins and Catechin Gallates isoflavone glycosides are initially hydrolyzed to their
TEAC (mM) E7 (V) Ep/2 (V) respective aglycons.62,63 However, glycosides are absorb-
catechin (3, 5, 7, 3′, 4′) 2.4 0.57 0.16 able, as recently proved by the LC-MS detection of quer-
epicatechin (3, 5, 7, 3′, 4′) 2.5 cetin 3-rutinoside in blood of volunteers after consumption
epigallocatechin (3, 5, 7, 3′, 4′, 5′) 3.8 0.42 of tomato purèe64 and of naringin (4′,5,7-trihydroxyfla-
epicatechin gallate 4.93 vanone-7-rhamnoglucoside) in urine of subjects who re-
epigallocatechin gallate 4.75 0.43 ceived orally naringin.65 In the case of catechins, epi-
Anthocyanidins gallocatechin gallate and epicatechin gallate have been
TEAC (mM) E7 (V) Ep/2 (V)
detected in human blood after intake of green tea, decaf-
feinated green tea extracts, and dark chocolate.66-69 Glu-
cyanidin (3, 5, 7, 3′, 4′) 4.4 -0.23 curonide and sulfate conjugates of (+)-catechin and 3′-O-
cyanidin 3-rutinoside 3.2
pelargonidin (3, 5, 7, 4′) 1.3
methyl-(+)-catechin have been determined in human plasma
after consumption of red wine.70
The percentage of absorption normally does not exceed
level of kaempferol (which differs from quercetin because a few percent of the ingested dose, as determined by
it has a lone hydroxyl group in ring B). These data confirm measuring the blood levels of intact flavonoids and their
further that the catechol structure in the B ring is the conjugates. Food composition may represent an important
major determinant for radical-scavenging capacity of the factor that affects bioavailability. Proteins may bind to
flavonoids. polyphenols,71 reducing their availability; by contrast,
In the case of isoflavones, the location of ring B at the alcohol may improve it,72 as evidenced by the increased
3-position of the heterocyclic ring greatly affects the radical- uptake of red wine phenolics as compared with levels
scavenging capacity, as determined by the TEAC assay resulting after the consumption of alcohol-free red wine.73
(Table 4). Thus, genistein is 2 times more potent than its In addition, recent data support an improved absorption
flavone relative apigenin. The single 4′-hydroxy group is of specific flavonoids in the presence of fats. Thus, catechins
required for scavenging, and on methoxylation, as in from green tea, oligomeric proanthocyanidins from grape
biochanin A, diminishes the potency. The 5,7-dihydroxy seeds, and silibinin from milk thistle are absorbed at higher
structure in ring A is also important, as evidenced by extent when administered as phospholipid complexes
comparing the pairs genistein/daidzein and biochanin rather than when free.74
A/formononetin. As for the other flavonoid classes, glyco- During absorption across the intestinal membrane, fla-
sylation negatively influences the radical-scavenging ca- vonols, flavones, isoflavones, and catechins are partly
pacity. transformed in their glucuronides and sulfates.75 Subse-
A different approach to evaluating the antioxidant quently, this small fraction of the absorbed flavonoids is
potential of flavonoids is based on their ability to increase metabolized by the liver enzymes, resulting in more polar
the resistance of isolated LDL to copper oxidation in vitro. conjugates being excreted in the urine or returned to the
This approach stems from the oxidative theory of athero- duodenum via the gallbladder. However, the major part of
genesis, which states that it is not LDL or VLDL that is ingested flavonoids is not absorbed and is largely degraded
atherogenic but the oxidized form of these lipoproteins.56 by the intestinal microflora. The bacterial enzymes catalyze
Indeed, several studies57-60 have shown that most of several reactions, including hydrolysis, cleavage of the
dietary flavonoids are effective against the oxidative modi- heterocyclic oxygen-containing ring, dehydroxylation, and
1040 Journal of Natural Products, 2000, Vol. 63, No. 7

Figure 4. Metabolic conversion of epigallocatechin gallate and rutin.

decarboxylation. Several phenolic acids are produced, cording to this approach and in agreement with other
depending on the structure of the flavonoid involved authors,80-83 we have proved that long-term consumption
(Figure 4).76 These phenolic acids can be reabsorbed and of green tea improves the levels of R-tocopherol in RBC (red
subjected to conjugation and O-methylation in the liver and blood cell) membranes and LDL. Plasma R-tocopherol and
then may enter into the circulation. This aspect is relevant hydrophilic antioxidant levels remained constant, while
for antioxidant protection, mainly for two reasons. The first β-carotene sligthly increased, possibly due to the protection
one is that phenolic acids may account for a large fraction from oxidation exerted by R-tocopherol.84 The content of
of the ingested flavonoids (30-60%), and the second is that polyunsaturated fatty acids in RBC membranes was im-
some of these acids, because of their catechol structure, proved, confirming that R-tocopherol, β-carotene, and cat-
possess a radical-scavenging ability comparable to that of echins (as co-antioxidants) act as effective lipid peroxida-
their intact precursors.77 This suggests that these metabo- tion inhibitors. These results suggest that long-term intake
lites may take part in the antioxidant protection, as of green tea guarantees a baseline plasma concentration
suggested by their TEAC values (Table 4).78 of catechins and their metabolites, which is able to induce
It is assumed that dietary flavonoids may display their an improvement of lipophilic vitamin levels. This modifica-
first antioxidant defense in the digestive tract, by limiting tion may be explained assuming that the antioxidant
ROS formation79 and scavenging them. Once absorbed, protection can be exerted through a cascade involving
either as aglycons and glycosides or, to a larger extent, as endogenous antioxidants, which react differently according
phenolic acids, they continue to exert an antioxidant effect, to their polarity and redox potential.85 More specifically,
although other systemic activities are possible. The in vivo flavonoids and their metabolites are capable of reducing
antioxidant effect may be evidenced by measuring the the highly oxidizing ROS, becoming less aggressive aroxyl
increase of the total antioxidant potential of plasma after radicals. Some of these aroxyl radicals (those with E7 >
a single (and, often, large) intake of flavonoid-containing 0.282 V) from their hydrophilic character may oxidize
food, beverages, or herbs and correlating this value to the ascorbate, which in turn is regenerated by glutathione. This
time course of plasma flavonoids. Alternatively, plasma could be the reason why, after a single dose of green tea
flavonoids (and/or their metabolites) and specific markers catechins either free or as phospholipid complexes, plasma
of the body antioxidant status (ascorbate, glutathione, ascorbate and total glutathione decrease transiently.74 This
R-tocopherol, β-carotene, polyunsaturated fatty acids, ma- decrease is concomitant with the time course of plasma
londialdehyde, 8-hydroxydeoxyguanosine) can be followed catechin (Cat-OH) concentration and the rise of plasma
during a long-term consumption of normal dosages. Ac- antioxidant capacity. On the basis of its redox potential,
Journal of Natural Products, 2000, Vol. 63, No. 7 1041

R-tocopherol (R-TOH, E7 ) 0.5V) could be oxidized by (5) Harman, D. J. Gerontology 1956, 11, 289-300.
(6) Halliwell, B.; Gutteridge, J. M. C. In Free Radicals in Biology and
radicals with E7 > 0.5 V (reaction 1), becoming a potential Medicine; Oxford University Press: Oxford, 1998.
pro-oxidant (reaction 2): (7) Warma, S. D.; Devamanoharan, P. S.; Morris, S. M. Crit. Rev. Food
Sci. Nutr. 1995, 35, 111-129.
R• + R-TOH f RH + R-TO•
(8) Wayner, D. D. M.; Burton, G. W.; Ingold, K. U.; Barclay, L. R. C.;
(1) Locke, S. J. Biochem. Biophys. Acta 1987, 924, 408-419.
(9) Halliwell, B. Lancet 1994, 344, 721-724.
R-TO• + LH(LDL) f R-TOH + L• f LOO• (2) (10) Sies, H. Antioxidants in Disease Mechanisms and Therapy, Advances
in Pharmacology, Vol. 38; Academic Press: San Diego, 1997.
Prevention of this pro-oxidant activity depends on the (11) Harborne, J. B. In Plant Flavonoids in Biology and Medicine; Cody,
V., Middleton, E., Harborne, J. B., Eds.; Alan R. Liss: New York,
rapid elimination of the R-tocopheroxyl radical (R-TO•).86 1986; pp 15-24.
This requires the presence of reductants (YH) capable of (12) Brouillard, R.; Cheminat, A. Prog. Clin. Biol. Res. 1988, 280, 93-
interacting with the lipophilic R-TO•: 106.
(13) Kuhnau, J. World Rev. Nutr. Diet. 1976, 24, 117-191.
(14) Pietta, P. G.; Mauri, P. L.; Simonetti, P.; Testolin, G. Fresenius J.
R-TO• + HY f R-TOH + Y• Anal. Chem. 1995, 352, 788-792.
(15) Herrman, K. J. Food Technol. 1976, 11, 433-448.
(16) Franke, A.; Custer, L. J.; Cerna, C. M.; Narala, K. K. J. Agric. Food
HY should have a strong reducing capacity (i.e., a redox Chem. 1994, 42, 1905-1913
potential lower than 0.5 V) and generate a harmless (17) Haslam, E. Plant Polyphenols; Cambridge University Press: Cam-
aqueous radical Y•, thereby preventing LOO• formation. A bridge, U.K., 1989.
(18) Mazza, G.; Miniati, E. Anthocyanins in Fruits, Vegetables and Grains;
variety of natural reductants can play this role, including CRC Press: Boca Raton, FL, 1993.
ascorbate, ubiquinol-10, some flavonoids, and phenolic (19) Pietta, P. G.; Simonetti, P. In Antioxidant Food Supplements in
acids (e.g., caffeic acid). Human Health; Packer, L., Hiramatsu, M., Yoshikawa, T., Eds.;
Academic Press: San Diego, 1999; pp 283-308.
Among flavonoids, quercetin and tea catechins (E7 0.22 (20) Harborne, J. B. In Flavonoids: Advances in Research Since 1986;
and ∼0.4 V, respectively) should be able to regenerate Harborne, J. B., Ed.; Chapman and Hall: London, 1994; pp 589-
R-tocopherol from the R-tocopheroxyl radical: (21) Shirley, B. W. Trends Plant Sci. 1996, 31, 377-382.
(22) Yamasaki, H.; Sakihama, Y.; Ikehara, N. Plant Physiol. 1997, 115,
R-TO• + Cat-OH f R-TO• + Cat-OH 1405-1412.
(23) Larson, R. A. Phytochemistry 1988, 27, 969-978.
(24) Pietta, P. G. In Flavonoids in Health and Disease; Rice-Evans, C. A.,
This may explain the improvement/maintainance of R-to- Packer, L., Eds.; Marcel Dekker: New York, 1998; pp 61-110.
copherol levels after intake of green tea catechins. It is (25) Ingram, D.; Sanders, K.; Kolybaba, M.; Lopez, M. Lancet 1997, 983,
quite conceivable that this sparing effect requires con- 990-994.
(26) Block, G.; Patterson, B.; Subar, A. Nutr. Cancer 1992, 17, 1-29.
sumption of green tea for a while, and during this period (27) Frei, B. C. Crit. Rev. Food Sci. 1995, 35, 83-97.
homeostatic recovery of the hydrophilic antioxidants ascor- (28) Gei, K. F. J. Nutr. Biochem. 1995, 6, 206-236.
(29) Gillman, M. W.; Cupples, L. A.; Gagnon, D.; Posner, B. M.; Ellison,
bate and glutathione takes place, as found experimen- C.; Castelli, W. P.; Wolf, P. J. Am. Med. Assoc. 1995, 273, 1113-
tally.85 1117.
Concerning the protection of LDL in vivo, the situation (30) Ness, A. R.; Powles, J. W. Int. J. Epidemiol. 1997, 6, 1-13.
(31) Peterson, J.; Dwyer, J. Nutr. Res. 1998, 12, 1995-2018
is far from conclusive. According to Vinson et al.,87 red wine (32) Gorinstein, S.; Barnitkowska, E.; Kulasek, G.; Zemser, M.; Tracht-
and black tea polyphenols are absorbed and protect both enberg, S. J. Nutr. 1998, 128, 2023-2027.
LDL and VLDL against oxidation by enrichment of lipo- (33) Wang, W.; Goodman, M. T. Nutr. Res. 1999, 19, 191-202.
(34) Hertog, M. G. L. Proc. Nutr. Soc. 1996, 55, 385-396.
proteins. However, some other in vivo studies have shown (35) Knekt, P.; Jarvinen, R.; Seppanen, R. Heliovaara, M.; Teppo, L.;
that ingestion of tea88,89 or red wine does not protect LDL Pukkala.; Aromaa, A. Am. J. Epidemiol. 1997, 146, 223-230.
against oxidation. The findings reinforce the hypothesis91 (36) Hertog, M. G. L.; Katan, M. B. In Flavonoids in Health and Disease;
Rice-Evans, C. A., Packer, L., Eds.; Marcel Dekker: New York, 1998;
that polyphenols are bound to plasma proteins other than pp 447-467.
LDL and, thus, are unable to preserve LDL. (37) Knekt, P.; Jarvinen, R.; Reunanen, A.; Maatela, J. Br. Med. J. 1996,
312, 478-481.
Conclusions. Dietary flavonoids represent an important (38) Rimm, E. B.; Katan, M. B.; Ascherio, A.; Stampfer, M.; Willet, W. C.
source of antioxidants, since their intake may reach 800 Ann. Int. Med. 1996, 125, 384-389.
mg/day. In the last years, many papers have been pub- (39) Hanasaki, Y.; Ogawa, S.; Fukui, S. Free Rad. Biol. Med. 1994, 16,
lished on the in vitro antioxidant activity of flavonoids, and (40) Ursini, F.; Maiorino, M.; Morazzoni, P.; Roveri, A.; Pifferi, G. Free
a correlation between the antioxidant capacity and chemi- Rad. Biol. Med. 1994, 16, 547-553.
cal structure has been assessed. However, the antioxidant (41) Korkina, L. G.; Afanas’ev, I. B. In Antioxidants in Disease Mechanisms
and Therapy; Sies, H., Ed.; Academic Press: San Diego, 1997; pp 151-
efficacy in vivo of flavonoids has been less thoroughly 163.
documented, possibly due to the limited knowledge on their (42) Brown, J. E.; Khodr, H.; Hider, R. C.; Rice-Evans, C. Biochem. J.
1998, 330, 1173-1178.
pharmacokinetics. Only recently has it been proved that a (43) Morris, C. J.; Earl, J. R.; Trenam, C. W.; Blake, D. R. Int. J. Biochem.
small fraction of the ingested dietary flavonoids is absorbed Cell Biol. 1995, 27, 109-122.
in either the aglycon or glycoside form, while the major (44) van Acker, S. A. B. E.; van den Berg, D. J.; Tromp, M. N. J. L.;
Griffoen, D. H.; van Bennekom, W. P.; van der Vijgh, W. J. F.; Bast,
part is extensively degraded to different phenolic acids. A. Free Rad. Biol. Med. 1996, 20, 275-283.
Both the absorbed flavonoids and their metabolites may (45) Jovanovic, S. V.; Steenken, S.; Tosic, M.; Marjanovic, B.; Simic, M.
display an in vivo antioxidant activity, which seems to G. J. Am. Chem. Soc. 1994, 116, 4846-4851.
(46) Buettner, G. R. Arch. Biochem. Biophys. 1993, 300, 535-543.
involve differently the physiological antioxidants, resulting (47) Robak, J.; Gryglewski, R. J. Biochem. Pharmacol. 1988, 37, 83-88.
in a sparing effect on R-tocopherol and β-carotene. (48) Hussain, S. R.; Cillard, J.; Cillard, P. Phytochemistry 1987, 26, 2489-
(49) Torel, J.; Cillard, J.; Cillard, P. Phytochemistry 1986, 25, 383-385.
Acknowledgment. The author wishes to acknowledge (50) McCord, J. M. Proc. Exp. Biol. Med. 1995, 202, 112-118.
Specchiasol s.r.l. (Verona, Italy) and Indena S.p.A. (Milan, (51) Cao, G.; Sofic, E.; Prior, R. L. Free Rad. Biol. Med. 1997, 22, 749-
Italy) for financial support. 760.
(52) Shadidi, F.; Janitha, P. K.; Wanasundara, P. D. Crit. Rev. Food Sci.
1992, 32, 67-103.
References and Notes (53) Rice-Evans, C. A.; Miller, N. J.; Paganga, G. Free Rad. Biol. Med.
(1) Davies, K. J. Biochem. Soc. Symp. 1995, 61, 1-31. 1996, 20, 933-956.
(2) Ames, B. N. Shigenaga, M. K.; Hagen, T. M. Proc. Natl. Acad. Sci. (54) van Acker, S. A. B. E.; van den Berg, D. J.; Tromp, M. N. J. L.;
U.S.A. 1993, 90, 7915-7922. Griffoen, D. H.; van der Vijgh, W. J. F.; Bast, A. Free Rad. Biol. Med.
(3) Halliwell, B. Nutr. Rev. 1997, 55, S44-S52. 1996, 20, 331-342.
(4) Blake, D.; Winyard, P. G. Immunopharmacology of Free Radical (55) Bors, W.; Heller, W.; Michel, C.; Saran, M. Methods Enzymol. 1990,
Species; Academic Press: San Diego, 1995. 186, 343-355.
1042 Journal of Natural Products, 2000, Vol. 63, No. 7

(56) Steinberg, D.; Parathasarathy, S.; Carew, T. E.; Khoo, J. C.; Witzum, (75) King, R. A.; Bursill, D. B. Am. J. Clin. Nutr. 1998, 67, 867-872.
J. L. N. Engl. J. Med. 1989, 320, 915-924. (76) Pietta, P. G.; Gardana, C.; Mauri, P. L. J. Chromatogr. 1997, 693,
(57) deWhalley, C. V.; Rankin, S. M.; Hoult, J. S. P.; Leake, D. S. Biochem. 249-255.
Pharmacol. 1990, 39, 1743-1750. (77) Merfort, I.; Heilmann, J.; Weiss, M.; Pietta, P. G.; Gardana, C. Planta
(58) Morel, I.; Lescoat, G.; Cogrel, P.; Cillard, J. Biochem. Pharmacol. Med. 1996, 62, 289-292.
1993, 45, 13-19. (78) Pietta, P. G.; Simonetti, P.; Gardana, C.; Mauri, P. L. J. Pharm.
(59) Igarashi, K.; Ohmuma, M. Biosci. Biotech. Biochem. 1995, 59, 595-
601. Biomed. Anal. 2000, in press.
(60) Vinson, J.; Dabbagh, Y. A.; Serry, M. M.; Jang, J. J. Agric. Food Chem. (79) Papas, A. M. Lipids 1996, 31, S77-S82.
1995, 43, 2800-2802. (80) Maffei-Facino, R.; Carini, A.; Aldini, G.; Calloni, M. T.; Bombardelli,
(61) Otero, P.; Viana, M.; Herrera, E.; Bonet, B. Free Rad. Res. 1997, 27, E.; Morazzoni, P. Planta Med. 1998, 64, 343-347.
619-626. (81) Brown, J. E.; Rice-Evans, C. A. Free Rad. Res. 1998, 29, 247-255.
(62) Manach, C.; Regerat, T.; Texier, O.; Agullo, G.; Demigne, C.; Remesy, (82) Fremont, L.; Gozzelino, M. T.; Franchi, M. P.; Linard, A. J. Nutr.
C. Nutr. Res. 1996, 16, 517-544. 1998, 128, 1495-1502.
(63) Nielsen, S. A.; Kall, M.; Justesen, U.; Schon, A.; Dragsted, L. O. (83) Pietta, P. G.; Simonetti, P.; Roggi, C.; Brusamolino, A.; Pellegrini,
Cancer Lett. 1997, 114, 173-174. N.; Maccarini, L.; Testolin, G. In Natural Antioxidants and Food
(64) Mauri, P. L.; Iemoli, L.; Gardana, C.; Riso, P.; Simonetti, P.; Porrini, Quality in Atherosclerosis and Cancer Prevention; Kumpulainen, J.
M.; Pietta. P. G. Rapid Commun. Mass Spectrom. 1999, 13, 924- T., Salonen, J. T., Eds.; Royal Society of Chemistry: Cambridge, UK,
(65) Ishii, K.; Furuta, T.; Kasuya, Y. J. Agric. Food Chem. 2000, 48, 56- 1996; pp 249-255.
59. (84) Palozza, P.; Krinsky, N. I. Arch. Biochem. Biophys. 1992, 297, 184-
(66) Pietta, P. G.; Simonetti, P.; Gardana, C.; Brusamolino, A.; Morazzoni, 187.
P.; Bombardelli, E. Biofactors 1998, 8, 111-118. (85) Pietta, P. G.; Simonetti, P. Biochem. Mol. Biol. Int. 1998, 44, 1069-
(67) Unno, T.; Kondo, K.; Itakura, H.; Takeo, T. Biosci. Biotech. Biochem. 1074.
1996, 60, 2066-2068. (86) Thomas, S. R.; Neuzil, J.; Mohr, D.; Stocker, R. Am. J. Clin. Nutr.
(68) Nakagawa, K.; Miyazawa, T. Anal. Biochem. 1997, 248, 41-49. 1995, 62, 1357S-1364S.
(69) Michelle, M.; Tavazzi, T.; Enslen, M.; Offord, E. A. Eur. J. Clin. Nutr. (87) Vinson, J. A.; Jang, J.; Dabbagh, Y. A.; Serry, M. M.; Cai, S. J. Agric.
1999, 53, 22-26. Food Chem. 1995, 43, 2798-2799.
(70) Bell, J. R. C.; Donovan, J. L.; Wong, R.; Waterhouse, A. L.; German, (88) van het Hof, K.; de Boer, H. S. M.; Wiseman, S. A.; Westrate, J. A.;
J. B.; Walzem, R. L.; Kasim-Karakas, S. E. Am. J. Clin. Nutr. 2000, Tijburg, L. B. M. Am. J. Clin. Nutr. 1997, 66, 1125-1132.
71, 103-108.
(89) McAnlis, G. T.; McEnemy, J.; Pearce, J.; Young, I. Y. Eur. J. Clin.
(71) Ichikawa, M.; Tsao, S. C.; Lin, T. H.; Miyahuchi, S.; Sawada, Y.; Iga,
T.; Hanano, M.; Sugiyama, Y. J. Hepatology 1992, 16, 38-49. Nutr. 1998, 52, 202-206.
(72) Serafini, M.; Maiani, G.; Ferro-Luzzi, A. J. Agric. Food Chem. 1997, (90) Caccetta, R. A.; Croft, K. D.; Beilin, L. J.; Puddey, I. B. Am. J. Clin.
45, 3148-3151. Nutr. 2000, 71, 67-74.
(73) Simonetti, P.; Gardana, C.; Mauri, P. L.; Pietta, P. G. Methods (91) McAnlis, G. T.; McEnemy, J.; Pearce, J.; Young, I. Y. Eur. J. Clin.
Enzymol. 2000, in press. Nutr. 1999, 53, 92-96.
(74) Pietta, P. G.; Simonetti, P.; Gardana, C.; Brusamolino, A., Morazzoni,
P.; Bombardelli, E. Biochem. Mol. Biol. Int. 1998, 46, 895-903. NP9904509